IOVS Journal of Cell Biology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

(Investigative Ophthalmology and Visual Science. 2006;47:2583-2588.)
© 2006 by The Association for Research in Vision and Ophthalmology, Inc.
DOI:  10.1167/iovs.05-1116
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nymark, S.
Right arrow Articles by Koskelainen, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nymark, S.
Right arrow Articles by Koskelainen, A.

A New Method for Measuring Free Drug Concentration: Retinal Tissue as a Biosensor

Soile Nymark,1 Charlotte Haldin,2 Heikki Tenhu,3 and Ari Koskelainen1

1From the Laboratory of Biomedical Engineering, Helsinki University of Technology, Espoo, Finland; and 2the Departments of Biological and Environmental Sciences and 3Chemistry, University of Helsinki, Helsinki, Finland.


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
PURPOSE. To develop a method of using isolated rat retina as a biosensor in experiments on controlled drug release for measuring the resultant concentration of free model drug in living tissue and for testing the biocompatibility of the polymers and polymeric nanostructures used as drug carriers.

METHODS. The method is based on the monotonic dependence of the photoresponse kinetics of retinal rods on the concentration of the membrane-permeable phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (IBMX). Changes in the time to peak (tp) of linear-range rod photoresponses were followed by transretinal ERG mass potential recordings in the aspartate-treated, dark-adapted rat retina. The dependence of tp on [IBMX] was measured, and the calibration curve thus obtained was used to determine the amount of IBMX released from polymeric structures. The biocompatibility of the carrier was first assessed by the degree to which rods retained stable function in the presence of the polymer or monomers alone.

RESULTS. The dependence of tp on [IBMX] was well-described by a second-order polynomial. After each change of [IBMX], a new equilibrium state was reached within 6 to 9 minutes, depending on temperature. The amounts of IBMX released from biocompatible polymeric structures were measurable with good accuracy in the range 10 to 300 µM.

CONCLUSIONS. This method enables accurate concentration determinations of the model drug IBMX in retinal tissue in drug-release experiments. The concentration dependence of the photoresponse kinetics has to be calibrated for each retina and temperature. The same preparation can be used for rapid testing of possible bioincompatibility of various molecules.

Drug release systems that can be controlled by stimuli external to the body may enable targeted drug delivery (i.e., the release of a drug both rapidly and locally), thereby reducing systemic effects. Polymeric materials and structures offer interesting possibilities for such controlled drug delivery.1 2 3 4 5 We studied the possibility of achieving rapid drug release from thermoresponsive polymeric nanostructures by local tissue heating. The polymeric structures we used (based on N-isopropylacrylamide, [NIPAAm] and vinylcaprolactame [VCa]) experience a sudden phase transition with a collapse in volume when heated above their lower critical solution temperature (LCST)—close to 32°C for both pure NIPAAm and VCa polymers. The temperature dependence of the phase transition and the rate of drug release can be readily measured in pure water and in physiological salt solutions with UV spectroscopy and light scattering, but these techniques do not allow determination of the amount of drug released in living tissue. For this purpose, we developed the method described herein.

The model tissue we used was the isolated rat retina. The retina is part of the brain, but compared with conventional brain slices, the retinal preparation is very stable for long periods (12 hours or more) and allows repeated temperature changes in the range of 20 °C to 42°C (which covers the LCSTs of NIPAAm and VCa). As our model drug, we used the membrane-permeable phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (IBMX). The determination of drug concentration in the retina is based on the fact that rod photoresponses decelerate monotonically with increasing IBMX concentration.6 The same preparation can also be used for rapid testing of the biocompatibility of the polymeric materials that are used to carry the drug.

In this report, we describe the method and give some examples of how the same preparation can be used both for preliminary biocompatibility tests and drug-release determinations. Although we used only IBMX as our model drug, any other molecules with well-known reversible effects on the phototransduction machinery and photoresponses could be used as well. These could be chosen to mimic drugs of interest with respect to particular properties (e.g., hydrophobicity and hydrophilicity).


   Methods
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Basis of the Method: Phototransduction in Rods and the Effect of IBMX
The light-sensitive cation channels in the rod outer segment (ROS) are kept open by internal transmitter molecules (cyclic guanosine monophosphate [cGMP]) bound to the channels. Absorption of a photon by a rhodopsin molecule in the disc membranes may activate several hundred G-protein (transducin) molecules, each of which in turn activates one molecule of cGMP-phosphodiesterase (PDE). Activated PDE may hydrolyze hundreds of cGMP molecules, leading to a decrease in the cytoplasmic cGMP concentration, release of cGMP from the channel binding sites, and closure of channels. This process leads to a decrease in the current influx into the cell, thereby generating the photoresponse.

The photoresponses to weak light pulses depend linearly on the intensity of the stimulus (i.e., the amplitude increases linearly with light intensity), whereas the kinetics stay unchanged. The membrane-permeable phosphodiesterase inhibitor IBMX reduces the amount of PDE that can be activated, thus increasing the time needed for an active G-protein to meet PDE and slowing down photoresponses. The kinetics of photoresponses to weak light may therefore be used as a measure of the IBMX concentration in the tissue.

Animals and Tissue Preparation
Wistar rats (Rattus norvegicus) were used in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Before the experiments the animals were dark-adapted overnight. They were first anesthetized by intraperitoneal injection of sodium pentobarbital (60 mg/kg) and then decapitated instantaneously. After enucleation and removal of the front part of the eye, the retina was gently detached from the eyecup and placed photoreceptor side up in a specimen-holder (schematic diagram in Fig. 1 ). The receptor side was superfused with Ringer’s solution at a constant flow rate throughout the experiment. The temperature of the retina was controlled by a brass heat exchanger below the Plexiglas specimen holder and monitored with a thermistor in the bath close to the retina.


Figure 1
View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 1. The setup for recording transretinal ERG.

 
Solutions and Chemicals
All experiments were conducted with Ringer’s solution containing (mM): Na+ 139.7, K+ 3.3, Mg2+ 2.0, Ca2+ 1.0; Cl– 143.2, glucose, 10.0; EDTA, 0.01; NaHCO3– 6; HEPES, 6; and sodium aspartate, 2. The solution was adjusted to pH 7.5 by NaOH at 25°C. BaCl2 (10 mM) was added in the lower electrode space to suppress glial currents.7 IBMX was purchased from Sigma. The monomers N-isopropylacrylamide (NIPAAm) and vinylcaprolactame (VCa) used for polymer synthesis as well as comonomer hexafluorobutylmethylacrylate (HFBMA) used for copolymerization with NIPAAm (to modify the LCST and to increase the hydrophobicity8 9 of the polymers) were all from Polysciences Inc. (Warrington, PA). The thermoresponsive polymers PNIPAAm and PVCa were synthesized and the latex particles prepared as described earlier.10

Electrical Recording and Stimulation
Transretinal ERG mass potentials were recorded as described previously.11 Stimuli were 20-ms flashes of nearly monochromatic 519 nm light (IL interference filter; Schott, Mainz, Germany). Stimulus intensity was controlled with a neutral-density wedge and filters. The signal was low-pass filtered at 100 Hz. The aspartate in the Ringer’s served to isolate the photoreceptor signal by disrupting synaptic transmission from photoreceptors to second-order neurons. Because of the low flash intensity and the rod dominance in the rat retina, cone responses may be neglected. Thus, our signals were practically pure rod signals up to their peak (tp, time to peak).

Analysis
Changes in photoresponse kinetics were characterized by the tp of linear-range photoresponses. The dependence of tp on [IBMX] was first measured at several concentrations of IBMX, and these calibration data were used to determine the amount of IBMX released from polymeric structures. In each solution photoresponses were recorded until a new steady state tp was achieved. In the release experiments, free [IBMX] was determined from the average tp of five photoresponses recorded in the equilibrium state. The error in [IBMX] was estimated by the SEM of the five tps, in both the calibration and the release experiments. In experiments at different temperatures, the tp versus [IBMX] relation was calibrated at each temperature.

Biocompatibility Tests
The amplitude and kinetics of rod photoresponses are very sensitive to physical or chemical changes in the environment of the photoreceptors. The physiological deterioration of rods is evident as decreasing amplitude of saturated photoresponses (i.e., responses to strong light pulses), decreasing sensitivity and slowing-down of photoresponses. A combination of these three properties can be used as a qualitative indicator of toxicity.


   Results
Top
 Abstract
 Methods
 Results
Discussion
 References
 
Rat Rod Photoresponses and the Stability of the Retina in ERG Recordings
The general characteristics of mass photoresponses from rat retina are illustrated by the response family in Figure 2 , covering a 5-log-unit range of light intensities starting from dim flashes that close only a low percentage of the light-sensitive channels. The three smallest responses are practically pure rod photoresponses, and they grow linearly with stimulus intensity while the kinetics remain unchanged. The responses to the strongest flashes display a saturated rod plateau preceded by a transient "nose" component. This component is likely to be of multiple origin, including cone currents as well as currents from voltage-sensitive channels in the rod inner segments.12 Such components are not present in the dim-flash photoresponses. Disregarding the "nose," the amplitude of the saturated rod response can be read in a consistent manner from the plateau.


Figure 2
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 2. ERG photoresponses recorded across the isolated, aspartate-treated rat retina. Stimuli were brief (20 ms) flashes of light of intensities increasing in 0.5-log-unit steps from 0.6 to 18,000 photons/µm2 at 20°C.

 
The saturating amplitude depends strongly on temperature7 as shown in Figure 3 . In this experiment (which lasted for ~12 hours) the temperature was changed several times between 12°C and 36°C. The steady state amplitudes at each temperature were very stable, as evident from the narrow error bars. Moreover, two separate epochs at 20°C yielded very similar saturating amplitudes, indicating good reversibility of the changes. The main conclusion from this and similar experiments is that the isolated retina is a remarkably stable preparation, which gives reproducible results, even in long experiments involving significant changes of temperature.


Figure 3
View larger version (14K):
[in this window]
[in a new window]
  		FIGURE 3. The amplitude of saturated rod photoresponses at different temperatures, plotted as a function of time from the onset of the experiment. The temperature was changed at ~1-hour intervals. The error bars are SEMs of the amplitude of three to five single responses measured at each temperature when a steady state had been reached. Note the high precision of the determination at each temperature as well as the similarity of the two measurements at 20°C, separated by ~5 hours. The stimulus intensity required to obtain a good saturation plateau increased from 750 photons/µm2 to 30 000 photons/µm2 between 12°C and 36°C.

 
Biocompatibility Tests
The polymeric materials may include some remnant monomers which, due to their smaller molecular size, are in fact more likely to be harmful than are the macromolecules. Therefore, it is important to test both the monomers and the polymers for biocompatibility. As examples, effects of NIPAAm and VCa monomers and polymers on the amplitudes of photoresponses to different stimulus intensities are shown in Figure 4 . NIPAAm monomers had either no discernible effect (two experiments) or slightly decreased photoresponse amplitudes (four experiments) at 10 mM concentration (Fig. 4A) . No decrease in amplitudes, however, was observed with PNIPAAm polymers (five experiments) (Fig. 4C) . At 10 mM concentration monomeric VCa clearly decreased photoresponse amplitudes (Fig. 4B) , but the photoresponses recovered completely after washout. When the same amount of VCa was introduced as polymers (average molar mass 106), a slight increase instead of decrease in photoresponse amplitudes was observed (Fig. 4D) .


Figure 4
View larger version (28K):
[in this window]
[in a new window]
  		FIGURE 4. Effects of polymers used as drug carriers and of the corresponding monomers on the amplitudes of rod responses to brief flashes of different intensities. The substances tested were 1 mM and 10 mM NIPAAm (A), 10 mM VCa (B), 10 mM PNIPAAm (C), and 10 mM PVCa (D). The stimulus intensities ranged from 3.4 ({diamondsuit}) to 340 photons/µm2 ({blacksquare}) in 0.5-log-unit steps (A), from 2.7 ({diamond}) to 270 photons/µm2 ({blacksquare}) in 0.4-log-unit steps (B), from 2.1 ({diamondsuit}) to 210 photons/µm2 ({blacksquare}) in 0.5-log-unit steps (C), and from 7.4 ({diamondsuit}) to 740 photons/µm2 ({blacksquare}) in 0.5-log-unit steps (D).

 
Calibration of the Dependence of Dim-Flash Photoresponse Kinetics on [IBMX]
The concentration-dependence of photoresponse kinetics on IBMX was determined in nine experiments in the range from 3 µM to 100 or 300 µM in ascending order of [IBMX]. The order was unimportant, however, as the effect of IBMX was completely reversible and repeatable.

The effect of IBMX on the kinetics of linear-range photoresponses is shown in Figure 5A . The amplitudes have all been normalized to 1 to make it easier to resolve changes in kinetics. IBMX clearly slowed down photoresponses in a concentration-dependent manner. In Figure 5B the time-behavior of tp in the course of the experiment is shown for the same retina. After each solution change, a new equilibrium state was achieved in 6 to 9 minutes, depending on temperature. In the whole concentration range tested, tp increased monotonically with increasing [IBMX]. After IBMX washout, tp returned to the original (pre-IBMX) level, indicating that the effect of IBMX is completely reversible. The behavior was similar in all experiments (n = 9).


Figure 5
View larger version (25K):
[in this window]
[in a new window]
  		FIGURE 5. (A) Linear-range photoresponses in normal Ringer’s (shortest tp) and in Ringer’s containing 3, 10, 30, and 100 µM IBMX (longest tp). Each trace is an average of six responses at 20°C. (B) Time-course of the perfusate switches and changes in tp in the same experiment. (C) Calibration curve for IBMX concentration as a function of tp extracted from the same data. The error bars are the SEM of the five equilibrium tp measured before the solution change. (D) The effect of temperature on the calibration curve. The data at 40°C ({blacksquare}) are from one retina and those at 20°C (•) and 30°C ({circ}) are from another (error bars as in C).

 
In Figure 5C the steady state tp (means of five responses ± SEM) from the same experiment are shown as a function of [IBMX]. The data in this as well as all other calibration experiments (n = 9) were well-fitted by a second-order polynomial: [IBMX] = b1 + b2 tp + b3 tp2. The factors b1, b2, and b3 were temperature-dependent and had to be determined separately for each retina and temperature. Calibrations from one retina or temperature cannot be transferred to another.

The effect of temperature on the shape of the calibration curve is shown in Figure 5D . The photoresponses accelerate with warming and thus the calibration curve moves toward lower tp, and the parameter b3 increases with rising temperature. However, the general (quadratic) form of the function holds at all temperatures tested (from 20 °C to 42°C).

Measurement of IBMX Release from Polymer Latices
The usefulness of the method is demonstrated in Figure 6 . The purpose of the experiment was to test how much of the IBMX (40 µM when translated to total concentration in the perfusate) loaded into PNIPAAm latex particles modified with HFBMA would be released at room temperature when the latex particles were dissolved in the perfusate. In Figure 6A , the effect of IBMX on photoresponse kinetics was first measured at three concentrations (10, 30, and 100 µM). The corresponding calibration function is shown in Figure 6B . After washout of IBMX and recovery of tp to the original value, perfusion was switched to a solution containing the latex particles loaded with IBMX. There was a clear increase of tp in the presence of the IBMX-loaded latex particles, although unloaded latex particles alone had no effect (Fig. 6A , inset). According to the calibration curve, the free IBMX concentration in the perfusate was approximately 25 ± 1 µM, corresponding to approximately two thirds of the total IBMX loaded into the latex particles.


Figure 6
View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 6. (A) Protocol for measurement of free [IBMX] in the tissue released from polymeric structures. First, the effect of 10, 30, and 100 µM IBMX on tp of linear range photoresponses was measured, and then the effect was determined of PNIPAAm latex particles modified with HFBMA and loaded with IBMX on tp. Inset: the effect of HFBMA modified PNIPAAm latex particles alone on tp. (B) Calibration curve for [IBMX] as function of tp and determination of [IBMX] released from the latex particles. Error bars: SEM of the five steady state tps measured at each IBMX concentration.

 

   Discussion
Top
 Abstract
 Methods
 Results
 Discussion
References
 
A central issue in stimulus-controlled drug release experiments is the measurement of the amount of drug released and the rate of release. Usually, the release measurements are based on conventional spectroscopic methods (e.g., UV spectroscopy on samples of the solution containing the drug released). However, if the free drug concentration is low, it may fall below the detection limit of the conventional methods. Further, if the spectral properties of the free and bound forms of the drug do not differ appreciably and if the drug-carrying vehicles are nanoparticles, it may be a formidable task to separate the vehicle from the solution rapidly enough to allow reliable measurement of the free drug concentration. The method described in this work enables accurate determinations of low concentrations of IBMX (or corresponding model drugs), in the case of IBMX from a few µM to hundreds of µM. The lower part of this range is well below the concentrations that can be measured with conventional methods.

In this study, IBMX was used as the model drug, because its effect on rod phototransduction and photoresponse kinetics is best characterized among the known rod phosphodiesterase (PDE6) inhibitors. Since its effect is intracellular, there is no risk that drug bound to the polymer particles could contribute to the measured effect.

Sandberg et al.13 have shown that IBMX increases the rod ERG b-wave implicit time in a concentration-dependent manner in the isolated perfused feline eye. Our study showed that IBMX concentration was reliably measurable in retinal tissue. Therefore, the system described herein may allow measurement of drug release from controlled release devices in vivo. Moreover, the method is not restricted to IBMX. Tissue concentrations of any other molecules that have reversible and concentration-dependent effects on photoresponse kinetics (acting on PDE or other phototransduction proteins) can be measured as well. The choice of the molecule, however, affects the concentration range that can be covered (e.g., with theophylline the range is from approximately 200 µM to 6 mM, approximately 1 log unit higher than with IBMX; Nymark and Koskelainen, unpublished data, 2006). A lower measurable concentration range can possibly be attained by PDE inhibitors with an inhibition constant (Ki) lower than with IBMX (Ki = 4490 nM for bovine rods14 ). Examples of this kind of molecule are vardenafil or sildenafil with Kis of 0.71 and 11 nM for bovine rods, respectively.14

In summary, the method has considerable advantages that appear useful, especially for drug-release experiments: (1) It measures concentrations in living tissue. In this respect, it is superior to most conventional techniques. (2) The accuracy of the method is high. We estimate that in the concentration range 10 to 300 µM of free IBMX, the error is ±10% or less in living tissue. Most of the variation is due to the possible inaccuracy in reading the linear-range tp and the subtle variability in the shape of the photoresponses. Although we had no means to make quantitative estimations by other techniques, we have confidence in the validity of these estimates. First, as the polymers themselves did not affect the tp of the photoresponses (e.g., see Fig. 6A ) and the effect of IBMX on tp was both completely reversible and repeatable, the changes in photoresponse kinetics most probably depended only on the free IBMX concentration in the perfusate. Second, when applied to polymeric structures shown to release all the IBMX loaded into them, the method yielded exactly the amount of IBMX that had been loaded.

The same preparation can be used for preliminary biocompatibility tests for various molecules. The retina preparation as a neural (brain) tissue is very sensitive to potentially hazardous molecules and materials. These biocompatibility tests are rapid and easy to carry out.


   Acknowledgements
 
The authors thank Kristian Donner for helpful comments on the manuscript.


   Footnotes
 
Supported by the Technology Development Centre (TEKES), Finland, by Grant 36154 from the Academy of Finland, and by the Oskar Öflund Foundation, Finland.

Submitted for publication August 23, 2005; revised January 17, 2006; accepted April 13, 2006.

Disclosure: S. Nymark, None; C. Haldin, None; H. Tenhu, None; A. Koskelainen, None

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Corresponding author: Soile Nymark, Laboratory of Biomedical Engineering, P.O. Box 2200, FI-02015 TKK, Finland; soile.nymark{at}hut.fi.


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Okano T eds. Biorelated Polymers and Gels: Controlled Release and Applications in Biomedical Engineering. 1998; Academic Press San Diego, CA.
  2. Kost J, Langer R. Responsive polymeric delivery systems. Adv Drug Deliver Rev. 2001;46:125–148.[Medline][Order article via Infotrieve]
  3. Hoffman AS. Hydrogels for biomedical applications. Adv Drug Deliver Rev. 2002;54:3–12.[CrossRef][Web of Science][Medline][Order article via Infotrieve]
  4. Kabanov AV, Batrakova EV, Sriadibhatla S, Yang Z, Kelly DL, Alakov VY. Polymer genomics: shifting the gene and drug delivery paradigms. J Control Release. 2005;101:259–271.[Medline][Order article via Infotrieve]
  5. de las Heras Alarcon C, Pennadam S, Alexander C. Stimuli responsive polymers for biomedical applications. Chem Soc Rev. 2005;34:276–285.
  6. Capovilla M, Cervetto L, Torre V. Antagonism between steady light and phosphodiesterase inhibitors on the kinetics of rod photoresponses. Proc Natl Acad Sci USA. 1982;79:6698–6702.[Abstract/Free Full Text]
  7. Nymark S, Heikkinen H, Haldin C, Donner K, Koskelainen A. Light responses and light adaptation in rat retinal rods at different temperatures. J Physiol. 2005;567:923–938.[Abstract/Free Full Text]
  8. Lowe TL, Virtanen J, Tenhu H. Hydrophobically modified responsive polyelectrolytes. Langmuir. 1999;15:4259–4265.[CrossRef]
  9. Lowe TL, Virtanen J, Tenhu H. Interactions of drugs and spin probes with hydrophobically modified polyelectrolyte hydrogels based on N-isopropylacrylamide. Polymer. 1999;40:2595–2603.[CrossRef]
  10. Lowe TL, Benhaddou M, Tenhu H. Partially fluorinated thermally responsive latices of linear and crosslinked copolymers. J Polym Sci Polym Phys. 1998;36:2141–2152.[CrossRef]
  11. Donner K, Hemilä S, Koskelainen A. Temperature-dependence of rod photoresponses from the aspartate-treated retina of the frog (Rana temporaria). Acta Physiol Scand. 1988;134:535–541.[Medline][Order article via Infotrieve]
  12. Green DG, Kapousta-Bruneau NV. A dissection of the electroretinogram from the isolated rat retina with microelectrodes and drugs. Vis Neurosci. 1999;16:727–741.[CrossRef][Web of Science][Medline][Order article via Infotrieve]
  13. Sandberg MA, Pawly BS, Crane WG, Schmidt SY, Berson EL. Effects of IBMX on the ERG of the isolated perfused cat eye. Vision Res. 1987;29:1421–1430.
  14. Zhang X, Feng Q, Cote RH. Efficacy and selectivity of phosphodiesterase-targeted drugs in inhibiting photoreceptor phosphodiesterase (PDE6) in retinal photoreceptors. Invest Ophthalmol Vis Sci. 2005;46:3060–3066.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nymark, S.
Right arrow Articles by Koskelainen, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nymark, S.
Right arrow Articles by Koskelainen, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS